Brett LindenbergEngagement rate is the share of an audience that took a defined action, out of the people who had the chance to take it, multiplied by 100. Each channel picks its own action and its own denominator. Social media tools count likes, comments, shares, and saves against followers, reach, or impressions. Google Analytics 4 counts engaged sessions against all sessions. For a training video, the useful version counts the learners who answered a question or made a choice at a given moment, divided by the learners who reached that moment. Name the action and the denominator before you compare two rates. In the short demo below, you work out one rate from 80 scenario answers across 200 qualified views.
How engagement rate is calculated on each channel
Every version has the same shape: the people or sessions that engaged, divided by the people or sessions that could have, multiplied by 100. The channels differ in what they count.
| Channel | Formula | What counts as engagement |
|---|---|---|
| Social media post (Hootsuite) | Engagements ÷ followers × 100; reach or impressions can replace followers | Likes, comments, shares, and saves |
| LinkedIn Page post | Interactions ÷ impressions | Clicks, reactions, comments, and shares |
| Website or app (Google Analytics 4) | Engaged sessions ÷ sessions × 100 | A session that lasts longer than 10 seconds, has a key event, or has 2 or more page or screen views |
| Hosted video (Wistia) | Average share of the video watched | Watching, which Wistia defines as how much of a video people watch on average |
| Interactive training video | Learners who acted at a moment ÷ learners who reached that moment × 100 | The answers, choices, and clicks the lesson asks for |
Sources: Hootsuite's guide to average engagement rates (June 2026 update), LinkedIn's help page on content analytics for your LinkedIn Page, Google Analytics Help on engagement rate and bounce rate, and Wistia's 2026 State of Video Report. No standards body defines the training video row; it is the definition we use and recommend.
Because the actions differ, rates from different rows do not compare. A 3% rate on a social post and a 40% rate on a training question measure different behavior. Google Analytics also defines bounce rate as the percentage of sessions that were not engaged, so a site's engagement rate and bounce rate add up to 100%. For engagement across a whole app or learning platform, such as how often people return, see our guide to user engagement metrics.
How to work out engagement rate
- Name one action. "Answered the scenario question" is countable. "Engaged with the video" is too vague to count.
- Choose the people who could take that action. For a question at 6:00, count the learners who reached 6:00, not everyone who pressed play. Leave out your own test and preview views.
- Count each person once. A learner who opens a resource link three times counts as one learner who acted.
- Divide the number who acted by the number who could have, and multiply by 100.
In the demo above, 80 of 200 qualified views include a scenario answer, so the rate is 80 ÷ 200 × 100 = 40%. The demo's two wrong answers show two calculation errors. Dividing the wrong way round (200 ÷ 80 × 100) gives 250%. Reporting the count of 80 answers as "80%" leaves out the denominator.
A third error is dividing all interactions by all viewers. If 100 learners produce 85 clicks and answers, calling that an 85% engagement rate overstates it when a few learners made most of those actions and the rest did nothing. Report the figure as interactions per view (0.85 in this case), and keep the rate for people.
A worked example for one lesson
Calculate the rate separately for each moment that asks learners to act. The figures below are illustrative.
| Moment | Learners who reached it | Learners who acted | Engagement rate |
|---|---|---|---|
| Required question at 1:30 | 200 | 200 | 100% |
| Optional job aid link at 3:10 | 184 | 46 | 25% |
| Optional scenario choice at 5:40 | 160 | 80 | 50% |
The required question reaches 100% by design, so it tells you nothing about engagement; read its correct answers instead. The two optional moments show how many learners chose to act. The falling "reached" column is a drop-off of 40 learners before 5:40, which is a separate problem to fix in the section before that point.
How do you measure the engagement and effectiveness of training videos?
Measure engagement with what learners do during the video, and effectiveness with whether they understood it and later use it at work. Four numbers for each video cover engagement and the first part of effectiveness:
| Measure | How to calculate it | What it shows |
|---|---|---|
| Watch rate | Average watched percentage across views | Whether learners stayed with the video |
| Drop-off point | The moment where many learners' furthest point stops | The section to fix first |
| Participation rate | Questions answered ÷ questions shown × 100, per learner, then averaged | Whether learners take part |
| First-attempt correct rate | Learners right on the first try ÷ learners who answered × 100, per question | Whether the explanation before the question worked |
The first three measure engagement. The fourth is your first evidence of effectiveness. The rest of the evidence sits outside the video, in error rates or call reviews from the weeks after training. Our guide to measuring training effectiveness covers that part, and how to measure content performance shows how to turn question results into a list of fixes.
Set the video up so these numbers mean something. If learners can drag the seek bar to the end, completion includes people who skipped the lesson. If a question is required, its participation rate is 100% for everyone who reaches it, so measure participation on optional moments and read correctness on required ones. In Mindstamp, watched percentage counts only the unique seconds a viewer played, divided by the video length, so skipping ahead and replaying do not inflate it. You can also stop learners from skipping ahead past a question they have not answered.
Where to find these numbers in Mindstamp
- The Views report shows watched percentage, participation, and correctness for each viewing session.
- The Questions report shows the answer breakdown for each question.
- The Interactions export lists every answer with its date and whether it was correct, so the earliest answer for each learner and question is the first attempt.
- Furthest point shows the latest moment each learner reached, which is how you find the drop-off.
Mindstamp defines participation and correctness per view: participation is the percentage of questions the viewer answered, and correctness is the percentage of answers that were correct. With Repeat Until Correct turned on, every attempt counts, so a wrong answer followed by a right one scores 50% for that question (passing scores and retries). Furthest point and last point explains the two position fields. Reports can be filtered by participation and correctness, which gives you the list of learners to follow up with. Each video's Reporting tab also has a drop-off graph.
What is a good engagement rate?
A good rate beats your own past content of the same type and length. Published figures help you judge a first result.
Social media
Hootsuite's benchmarks, updated in June 2026, call an engagement rate between 1% and 5% "good" and list these average rates per post: 3.5% on Instagram, 3.4% on LinkedIn, 1.8% on X, 1.5% on TikTok, and 1.3% on Facebook. Hootsuite does not say which denominator those averages use, so compare them with your own rate only when you calculate it the same way each time. Hootsuite also notes that smaller accounts often see higher rates than large ones.
Training videos
We know of no published industry benchmark for interaction rates in training video. The closest public data comes from a 2014 study by Philip Guo, Juho Kim, and Rob Rubin of 6.9 million video sessions in four edX courses. The researchers measured engagement in two ways: how long students watched, and whether they attempted the assessment problem that followed a video within 30 minutes. Students could move on without attempting it. The share of sessions followed by a problem attempt fell from 56% for the shortest videos to 31% for the longest. Compare those figures only with optional moments in your own videos, because a required question reaches 100% participation by design.
A raw count of actions per view mixes what learners do with how many questions and buttons each video offers. Across 8.07 million views of 18,315 Mindstamp videos (2018 to October 2026), the average view included 2.14 viewer actions (answers, clicks, and form entries). Views of videos of 1 minute or less averaged 0.58 actions, and views of videos of 10 to 20 minutes averaged 6.70 actions. Many factors differ between those groups, including how many interactions each video contains, so compare rates per opportunity, as in the worked example above. These figures describe Mindstamp videos, which are mostly training, marketing, and sales videos, and show a pattern, not a benchmark for all video. For watch-rate figures by video length, see our guide to video engagement metrics.
Engagement rate vs completion rate
Completion rate is the share of views that reach the end or meet your completion rule. Engagement rate is the share of learners who took part along the way. A muted video in a background tab can complete, but it cannot answer a question. Read the two together:
| What you see | Likely cause | What to try |
|---|---|---|
| High completion, low participation | Learners let the video play without working through it | Put a required question at each key concept |
| Low completion, high participation among learners who stay | One long or hard section loses people | Find the drop-off and shorten or split that section |
| High completion and participation, low first-attempt correct rate | Learners take part, but the explanation does not land | Rewrite the segment before the most-missed question |
How to raise engagement in a training video
Start with the steepest drop-off, because no interaction placed after it reaches the learners who left. Then work through these changes one at a time:
- Ask a question right after each key concept. In a 2013 study in PNAS, Szpunar, Khan, and Schacter found across two experiments that short memory tests between segments of an online lecture reduced mind wandering, increased note-taking, and improved learning. A Mindstamp question pauses the video and can require an answer before playback continues.
- Split long videos into shorter parts. In the edX study above, median engagement time was at most 6 minutes regardless of video length, and students often watched less than half of videos longer than 9 minutes. The researchers recommended chunks of less than 6 minutes.
- Make the action part of the lesson. A branching scenario that asks "What would you do next?" and plays the result of the learner's choice gives practice, while a "click to continue" button records only a click.
- Show a resource at the moment the narrator mentions it. A button or hotspot that opens the job aid at that point puts it in front of learners while the step is on screen.
Engagement rate for marketing video
When a video supports a campaign or a sales page, engagement is a step toward the result you report. Count the action that matters to the campaign, such as clicks on "Book a demo" ÷ viewers who reached the button, then follow those viewers to the conversion. For the page that holds the video, Mick-Mar's 2025 guide to improving website conversion rates covers baselines and A/B tests, and one of its sample test ideas is a product video against a static photo. For page and article metrics, see our guide to content engagement metrics.
Common questions about engagement rate
Can engagement rate be higher than 100%?
Yes, on social media. When followers are the denominator and a post reaches people who do not follow you, engagements can exceed followers. A rate can also pass 100% when you count actions instead of people. For a training video, divide learners who acted by learners who reached the moment, and the rate stays between 0% and 100%.
Is a higher engagement rate always better?
No. Many replays of one step can mean the step is confusing. A required question pushes its own rate to 100%. A correct rate near 100% on every question can mean the questions are too easy to show anything. In Mindstamp, engagement time runs from the start of a session to its last activity and includes pauses, so a long engagement time can also mean a learner left the tab open. Read engagement next to first-attempt correctness and drop-off.
What is the ideal length for an engaging training video?
The edX research points to segments of less than 6 minutes, measured on online course videos. The Mindstamp platform data cited above does not show a straight line: videos of 2 to 5 minutes averaged 39.2% watched, and videos of 10 to 20 minutes averaged 53.8% watched. Many factors differ between those groups. Cut a video where the topic allows a natural break, and check where your learners stop before you shorten anything.
Do I need special software to measure training video engagement?
You need a player that records answers and positions for each learner. An LMS records only what the course sends it, which for many video packages is completion and a score. Interactive video software such as Mindstamp records each answer and click against the learner who made it, on top of video you already host. On supported plans, a Mindstamp SCORM package also sends completion and a score to your LMS.
To start, pick one training video, write down the one action that shows a learner took part, and calculate its rate for the last group of learners. Then fix the section before the moment with the lowest rate, and compare the next group with the last one.
Sources and further reading
- Hootsuite, average engagement rates for 12 industries (June 2026 update).
- Google Analytics Help, engagement rate and bounce rate.
- Guo, Kim, and Rubin, How video production affects student engagement, Learning at Scale conference, 2014.
- Szpunar, Khan, and Schacter, Interpolated memory tests reduce mind wandering and improve learning of online lectures, PNAS, 2013.
